CN112783688A - Erasure code data recovery method and device based on available partition level - Google Patents
Erasure code data recovery method and device based on available partition level Download PDFInfo
- Publication number
- CN112783688A CN112783688A CN202110184248.8A CN202110184248A CN112783688A CN 112783688 A CN112783688 A CN 112783688A CN 202110184248 A CN202110184248 A CN 202110184248A CN 112783688 A CN112783688 A CN 112783688A
- Authority
- CN
- China
- Prior art keywords
- node
- recovery
- data
- target
- data block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/08—Error detection or correction by redundancy in data representation, e.g. by using checking codes
- G06F11/10—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
- G06F11/1004—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's to protect a block of data words, e.g. CRC or checksum
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/08—Error detection or correction by redundancy in data representation, e.g. by using checking codes
- G06F11/10—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
- G06F11/1008—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices
- G06F11/1048—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices using arrangements adapted for a specific error detection or correction feature
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Security & Cryptography (AREA)
- Memory System Of A Hierarchy Structure (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses an erasure code data recovery method and device based on available partition levels, wherein the method comprises the following steps: acquiring a data recovery request; responding to the data recovery request, and determining the node fault type if a node fault occurs in the data recovery process; selecting a corresponding target recovery strategy according to the node fault type; and recovering the node data corresponding to the node fault according to the target recovery strategy, and realizing the recovery of the lost data block of the first node. By implementing the method and the device, the problem of node failure in the data recovery process can be effectively solved, the occupied bandwidth of data recovery is reduced, network delay is reduced, the data recovery efficiency can be accelerated, and the data recovery performance is improved.
Description
Technical Field
The present application relates to the field of cloud storage technologies, and in particular, to an erasure code data recovery method and apparatus based on available partition levels.
Background
Modern data storage systems use erasure codes to provide high reliability and low storage overhead. The main problem of current erasure coding techniques is that high network traffic is generated when a recovery node fails. The recently developed usable area az (availability zones) -Recovery erasure code architecture can find the best Recovery path to reduce the bandwidth occupancy under various network configuration situations. However, as the capacity of the storage system is larger and larger, the probability of node failure is more and more frequent, so that more and more failures occur in the data recovery process.
Therefore, how to provide a more efficient data recovery scheme when a node failure occurs in the data recovery process is a problem that needs to be solved at present.
Disclosure of Invention
In order to overcome the defects in the prior art, the present application aims to provide an erasure code data recovery method and apparatus based on an available partition AZ level, which can reduce network delay and improve data recovery performance, thereby accelerating data recovery efficiency.
To achieve the above and other objects, the present application provides an erasure code data recovery method based on available partition levels, comprising the steps of:
acquiring a data recovery request, wherein the data recovery request is used for requesting to recover a lost data block of a first node;
responding to the data recovery request, and determining the node fault type if a node fault occurs in the data recovery process;
selecting a corresponding target recovery strategy according to the node fault type;
and recovering the node data corresponding to the node fault according to the target recovery strategy, and realizing the recovery of the lost data block of the first node.
Optionally, the data recovery request is specifically configured to request to read a target data block stored in a second node from a target partition, so as to implement data recovery of the first node based on the read target data block; and supports data recovery across available partitions or the same available partition.
Optionally, if the node failure type is a master node failure, and the target recovery policy is a first recovery policy, the recovering the node data corresponding to the node failure according to the target recovery policy, and implementing recovery of the lost data block of the first node includes:
after the cloud storage system pulls up a new main node, reading a target data block stored in a second node from a target partition according to the indication of the data recovery request again, and storing the read target data block in a data cache module;
and performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node.
Optionally, if the node failure type is a cached node failure, and the target recovery policy is a second recovery policy, the recovering the node data corresponding to the node failure according to the target recovery policy, and implementing recovery of the lost data block of the first node includes:
reading a target data block stored in a second node from a target partition according to the indication of the data recovery request, and storing the read target data block in a data cache module;
according to the indication of the second recovery strategy, obtaining a target data block of a cached node from the data caching module, and writing the target data block of the cached node into a cached node which is newly pulled up by the cloud storage system, so as to realize data recovery of the cached node;
performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node;
the cached node failure is used for indicating that the cached node fails after a main node caches a target data block of the cached node in the data recovery process; the cached node is any one of the second nodes.
Optionally, if the node failure type is an uncached node failure, and the target recovery policy is a third recovery policy, the recovering the node data corresponding to the node failure according to the target recovery policy, and implementing recovery of the lost data block of the first node includes:
generating new data recovery requests for the first node and the uncached node according to the indication of the third recovery strategy;
responding to the new data recovery request, and realizing data recovery of the first node and the uncached node;
the uncached node failure is used for indicating that the uncached node fails when a main node does not finish caching a target data block of the uncached node in a data recovery process; the uncached node is any one of the second nodes.
To achieve the above and other objects, the present application also provides an erasure code data recovery apparatus based on available partition levels, including:
an obtaining unit, configured to obtain a data recovery request, where the data recovery request is used to request recovery of a lost data block of a first node;
the determining unit is used for responding to the data recovery request, and determining the node fault type if a node fault occurs in the data recovery process;
the selection unit is used for selecting a corresponding target recovery strategy according to the node fault type;
and the recovery unit is used for recovering the node data corresponding to the node failure according to the target recovery strategy and realizing the recovery of the lost data block of the first node.
Optionally, the data recovery request is specifically configured to request to read a target data block stored in a second node from a target partition, so as to implement data recovery of the first node based on the read target data block; and supports data recovery across available partitions or the same available partition.
Optionally, the node failure type is a master node failure, the target restoration policy is a first restoration policy, and the restoration unit is specifically configured to:
after the cloud storage system pulls up a new main node, reading a target data block stored in a second node from a target partition according to the indication of the data recovery request again, and storing the read target data block in a data cache module;
and performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node.
Optionally, the node failure type is a cached node failure, the target restoration policy is a second restoration policy, and the restoration unit is specifically configured to:
reading a target data block stored in a second node from a target partition according to the indication of the data recovery request, and storing the read target data block in a data cache module;
according to the indication of the second recovery strategy, obtaining a target data block of a cached node from the data caching module, and writing the target data block of the cached node into a cached node which is newly pulled up by the cloud storage system, so as to realize data recovery of the cached node;
performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node;
the cached node failure is used for indicating that the cached node fails after a main node caches a target data block of the cached node in the data recovery process; the cached node is any one of the second nodes.
Optionally, the node failure type is an uncached node failure, the target recovery policy is a third recovery policy, and the recovery unit is specifically configured to:
generating new data recovery requests for the first node and the uncached node according to the indication of the third recovery strategy;
responding to the new data recovery request, and realizing data recovery of the first node and the uncached node;
the uncached node failure is used for indicating that the uncached node fails when a main node does not finish caching a target data block of the uncached node in a data recovery process; the uncached node is any one of the second nodes.
It can be seen from the above that the present application provides an erasure code data recovery method and apparatus based on available partition levels, which can achieve the following beneficial effects: the problem of node failure in the data recovery process can be effectively solved, the bandwidth occupied by data recovery is reduced, network delay is reduced, the data recovery efficiency can be accelerated, and the data recovery performance is improved.
Drawings
Fig. 1 is a schematic flowchart of an erasure code data recovery method based on available partition levels according to an embodiment of the present application.
Fig. 2 is a schematic structural diagram of an erasure code data recovery apparatus based on available partition levels according to an embodiment of the present application.
Detailed Description
Other advantages and effects of the present application will become apparent to those skilled in the art from the present disclosure, which is made apparent from the following detailed description of the embodiments given by way of example only and taken in conjunction with the accompanying drawings. The present application is capable of other and different embodiments and its several details are capable of modification in various other respects, all without departing from the spirit and scope of the present application.
The method and the device for providing the optimal data Recovery performance when the node faults occur in the data Recovery process are designed on the basis of researching how the AZ-Recovery architecture is more effective in dealing with the node faults occurring in the data Recovery process and investigating and analyzing the existing strategy of how to reduce network delay and accelerate the data Recovery process when different node faults occur.
Please refer to fig. 1, which is a flowchart illustrating an erasure code data recovery method based on available partition levels according to an embodiment of the present application. The method as shown in fig. 1 comprises the following implementation steps:
s101, a data recovery request is obtained, and the data recovery request is used for requesting to recover the lost data block of the first node.
After any node in the cloud storage system of the available partition fails, the cloud storage system can pull up a new node to replace the original node and initiate a corresponding data recovery request to recover the lost data in the original node, so as to continue to provide corresponding service.
The available partitions referred to herein have a plurality of nodes deployed therein, one of which is a master node (master node), and the remaining nodes may be referred to as slave nodes. The nodes may include, but are not limited to, servers, computers, and other network devices.
Specifically, the data recovery request initiated by any first node in the cloud storage system may be obtained, where the data recovery request may be used to request to recover a lost data block of the first node, and specifically, may be used to request to read a target data block stored in a second node from a target partition, so as to facilitate subsequent recovery of the lost data block in the first node based on the read target data block. The target partition may be an available partition where the first node is located, or may also be another available partition, in other words, the present application supports data reading and recovery of the same available partition or across available partitions.
And S102, responding to the data recovery request, and determining the node fault type if a node fault occurs in the data recovery process.
In the data recovery process in response to the data recovery request, a (node) fault may occur, and at this time, the corresponding node fault type needs to be determined. There are generally two main categories: a master node failure and a node failure of a cache node. If it is the master node that fails during the data recovery process, all the data blocks it manages caching will be lost, as well as data blocks from other partitions that may be being transferred. A cached node fails if other failures involve an already cached node (which may be referred to as a cached node), i.e., the home node has read the corresponding data block from the cached node. The newly pulled node of the system can immediately be recovered by copying the corresponding data block from the master node. If other failures involve an uncached node (which may be referred to as an uncached node), i.e., the home node has not read the corresponding data block from the uncached node, the uncached node fails at this time. Then, the corresponding data recovery strategy needs to be redesigned, a new data recovery request is generated, and the data blocks in the corresponding nodes are recovered according to the new data recovery request.
S103, selecting a corresponding target recovery strategy according to the node fault type.
S104, according to the target recovery strategy, recovering the node data corresponding to the node fault, and realizing the recovery of the lost data block of the first node.
As a possible implementation manner, if the node failure type is a master node failure, and the target recovery policy is a first recovery policy, according to an indication of the first recovery policy, after the cloud storage system pulls up a new master node, the master node responds to the data recovery request again, and reads a target data block stored in a second node from a target (available) partition, where the target partition may be an available partition where the master node is located, or may be another partition, and the number of the target partitions is not limited, and is set according to an actual requirement. Further optionally, the master node may store the read target data block in its own data cache module, so as to facilitate subsequent use. Then, the master node may perform recovery calculation on the read target data block to recover and obtain the lost data block of the first node, and then rewrite the lost data block into the first node, thereby implementing data loss recovery of the first node.
As another possible implementation, if the node failure type is a cached node failure and the target recovery policy is a second recovery policy, it may be determined that the data block of the failed cached node is cached in the master node. Correspondingly, after the cloud storage system pulls up the new cached node, the new cached node can directly copy the corresponding data block stored before from the data caching module of the main node, so that the data recovery of the cached node is realized. In addition, since the master node does not fail in this case, the master node may continue to respond to the data recovery request, read the target data block stored in the second node from the target partition, and write the read target data block into the data cache module. And then, performing recovery calculation on the read target data block to obtain a lost data block of the first node. And finally, writing the recovered lost data block into the first node, thereby completing the data recovery of the first node.
As another possible implementation manner, if the node failure type is an uncached node failure, the target recovery policy is a third recovery policy, and since the target data block of the uncached node is not completely cached in the master node at this time, the data recovery of the first node cannot be realized by using the target data block. Therefore, the present application may regenerate new data recovery requests for the first node and the uncached node according to the indication of the third recovery policy, where the new data recovery requests are used to request recovery of the missing data blocks in the first node and the uncached node. In practical applications, the new data recovery request may be specifically used to request to read the data block stored in the target node from the new available partition, so as to facilitate the subsequent data recovery of the corresponding node based on the read data block, and for how to implement the data recovery, reference may be specifically made to the foregoing related description, and details are not repeated here.
Through implementing the embodiment of the application, compared with the prior art, the application carries out corresponding research on the node faults in the data recovery process, considers various conditions of the node faults in the data recovery process, classifies and summarizes the node faults and provides corresponding data recovery strategies, and then the data recovery of the corresponding nodes can be realized more efficiently according to the corresponding data recovery strategies.
Fig. 2 is a schematic structural diagram of an erasure code data recovery apparatus based on available partition levels according to an embodiment of the present application. The apparatus shown in fig. 2 comprises an acquisition unit 201, a determination unit 202, a selection unit 203 and a recovery unit 204. Wherein the content of the first and second substances,
the obtaining unit 201 is configured to obtain a data recovery request, where the data recovery request is used to request to recover a lost data block of a first node;
the determining unit 202 is configured to, in response to the data recovery request, determine a node failure type if a node failure occurs in a data recovery process;
the selecting unit 203 is configured to select a corresponding target recovery policy according to the node fault type;
the recovering unit 204 is configured to recover, according to the target recovery policy, the node data corresponding to the node failure, and implement recovery of the lost data block of the first node.
Optionally, the data recovery request is specifically configured to request to read a target data block stored in a second node from a target partition, so as to implement data recovery of the first node based on the read target data block; and supports data recovery across available partitions or the same available partition.
Optionally, the node failure type is a master node failure, the target restoration policy is a first restoration policy, and the restoration unit 204 is specifically configured to:
after the cloud storage system pulls up a new main node, reading a target data block stored in a second node from a target partition according to the indication of the data recovery request again, and storing the read target data block in a data cache module;
and performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node.
Optionally, the node failure type is a cached node failure, the target restoration policy is a second restoration policy, and the restoration unit 204 is specifically configured to:
reading a target data block stored in a second node from a target partition according to the indication of the data recovery request, and storing the read target data block in a data cache module;
according to the indication of the second recovery strategy, obtaining a target data block of a cached node from the data caching module, and writing the target data block of the cached node into a cached node which is newly pulled up by the cloud storage system, so as to realize data recovery of the cached node;
performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node;
the cached node failure is used for indicating that the cached node fails after a main node caches a target data block of the cached node in the data recovery process; the cached node is any one of the second nodes.
Optionally, the node failure type is an uncached node failure, the target recovery policy is a third recovery policy, and the recovery unit 204 is specifically configured to:
generating new data recovery requests for the first node and the uncached node according to the indication of the third recovery strategy;
responding to the new data recovery request, and realizing data recovery of the first node and the uncached node;
the uncached node failure is used for indicating that the uncached node fails when a main node does not finish caching a target data block of the uncached node in a data recovery process; the uncached node is any one of the second nodes.
Through implementing the embodiment of the application, compared with the prior art, the application carries out corresponding research on the node faults in the data recovery process, considers various conditions of the node faults in the data recovery process, classifies and summarizes the node faults and provides corresponding data recovery strategies, and then the data recovery of the corresponding nodes can be realized more efficiently according to the corresponding data recovery strategies.
The above embodiments are merely illustrative of the principles and utilities of the present application and are not intended to limit the application. Modifications and variations can be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the present application. Therefore, the protection scope of the present application should be as set forth in the claims.
Claims (10)
1. An erasure code data recovery method based on available partition levels, comprising:
acquiring a data recovery request, wherein the data recovery request is used for requesting to recover a lost data block of a first node;
responding to the data recovery request, and determining the node fault type if a node fault occurs in the data recovery process;
selecting a corresponding target recovery strategy according to the node fault type;
and recovering the node data corresponding to the node fault according to the target recovery strategy, and realizing the recovery of the lost data block of the first node.
2. The method according to claim 1, wherein the data recovery request is specifically configured to request a target data block stored in the second node to be read from a target partition, so as to implement data recovery of the first node based on the read target data block; and supports data recovery across available partitions or the same available partition.
3. The method according to claim 2, wherein the node failure type is a master node failure, the target recovery policy is a first recovery policy, and if the node data corresponding to the node failure is recovered according to the target recovery policy and the recovery of the lost data block of the first node is implemented, the method includes:
after the cloud storage system pulls up a new main node, reading a target data block stored in a second node from a target partition according to the indication of the data recovery request again, and storing the read target data block in a data cache module;
and performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node.
4. The method according to claim 2, wherein the node failure type is a cached node failure, the target recovery policy is a second recovery policy, and if the node data corresponding to the node failure is recovered according to the target recovery policy and the recovery of the lost data block of the first node is implemented, the recovering includes:
reading a target data block stored in a second node from a target partition according to the indication of the data recovery request, and storing the read target data block in a data cache module;
according to the indication of the second recovery strategy, obtaining a target data block of a cached node from the data caching module, and writing the target data block of the cached node into a cached node which is newly pulled up by the cloud storage system, so as to realize data recovery of the cached node;
performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node;
the cached node failure is used for indicating that the cached node fails after a main node caches a target data block of the cached node in the data recovery process; the cached node is any one of the second nodes.
5. The method according to claim 2, wherein the node failure type is an uncached node failure, the target recovery policy is a third recovery policy, and if the node data corresponding to the node failure is recovered according to the target recovery policy and the recovery of the lost data block of the first node is implemented, the recovering includes:
generating new data recovery requests for the first node and the uncached node according to the indication of the third recovery strategy;
responding to the new data recovery request, and realizing data recovery of the first node and the uncached node;
the uncached node failure is used for indicating that the uncached node fails when a main node does not finish caching a target data block of the uncached node in a data recovery process; the uncached node is any one of the second nodes.
6. An erasure code data recovery apparatus based on available partition levels, comprising:
an obtaining unit, configured to obtain a data recovery request, where the data recovery request is used to request recovery of a lost data block of a first node;
the determining unit is used for responding to the data recovery request, and determining the node fault type if a node fault occurs in the data recovery process;
the selection unit is used for selecting a corresponding target recovery strategy according to the node fault type;
and the recovery unit is used for recovering the node data corresponding to the node failure according to the target recovery strategy and realizing the recovery of the lost data block of the first node.
7. The erasure code data recovery apparatus based on available partition level according to claim 6, wherein the data recovery request is specifically configured to request to read a target data block stored in a second node from a target partition, so as to implement data recovery of the first node based on the read target data block; and supports data recovery across available partitions or the same available partition.
8. The apparatus according to claim 7, wherein the node failure type is a master node failure, the target recovery policy is a first recovery policy, and the recovery unit is specifically configured to:
after the cloud storage system pulls up a new main node, reading a target data block stored in a second node from a target partition according to the indication of the data recovery request again, and storing the read target data block in a data cache module;
and performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node.
9. The apparatus according to claim 7, wherein the node failure type is a cached node failure, the target recovery policy is a second recovery policy, and the recovery unit is specifically configured to:
reading a target data block stored in a second node from a target partition according to the indication of the data recovery request, and storing the read target data block in a data cache module;
according to the indication of the second recovery strategy, obtaining a target data block of a cached node from the data caching module, and writing the target data block of the cached node into a cached node which is newly pulled up by the cloud storage system, so as to realize data recovery of the cached node;
performing recovery calculation on the read target data block to obtain the lost data block, and writing the lost data block into the first node;
the cached node failure is used for indicating that the cached node fails after a main node caches a target data block of the cached node in the data recovery process; the cached node is any one of the second nodes.
10. The apparatus according to claim 7, wherein the node failure type is an uncached node failure, the target recovery policy is a third recovery policy, and the recovery unit is specifically configured to:
generating new data recovery requests for the first node and the uncached node according to the indication of the third recovery strategy;
responding to the new data recovery request, and realizing data recovery of the first node and the uncached node;
the uncached node failure is used for indicating that the uncached node fails when a main node does not finish caching a target data block of the uncached node in a data recovery process; the uncached node is any one of the second nodes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110184248.8A CN112783688B (en) | 2021-02-10 | 2021-02-10 | Erasure code data recovery method and device based on available partition level |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110184248.8A CN112783688B (en) | 2021-02-10 | 2021-02-10 | Erasure code data recovery method and device based on available partition level |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112783688A true CN112783688A (en) | 2021-05-11 |
| CN112783688B CN112783688B (en) | 2022-06-03 |
Family
ID=75761519
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110184248.8A Active CN112783688B (en) | 2021-02-10 | 2021-02-10 | Erasure code data recovery method and device based on available partition level |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112783688B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114595092A (en) * | 2022-04-28 | 2022-06-07 | 阿里云计算有限公司 | Distributed storage system, data reconstruction method, device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106662983A (en) * | 2015-12-31 | 2017-05-10 | 华为技术有限公司 | Method, apparatus and system for data reconstruction in distributed storage system |
| CN107544862A (en) * | 2016-06-29 | 2018-01-05 | 中兴通讯股份有限公司 | A kind of data storage reconstructing method and device, memory node based on correcting and eleting codes |
| US20180060169A1 (en) * | 2016-08-26 | 2018-03-01 | Netapp, Inc. | Multiple node repair using high rate minimum storage regeneration erasure code |
| CN111818124A (en) * | 2020-05-29 | 2020-10-23 | 平安科技（深圳）有限公司 | Data storage method, data storage device, electronic equipment and medium |
-
2021
- 2021-02-10 CN CN202110184248.8A patent/CN112783688B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106662983A (en) * | 2015-12-31 | 2017-05-10 | 华为技术有限公司 | Method, apparatus and system for data reconstruction in distributed storage system |
| CN107544862A (en) * | 2016-06-29 | 2018-01-05 | 中兴通讯股份有限公司 | A kind of data storage reconstructing method and device, memory node based on correcting and eleting codes |
| US20180060169A1 (en) * | 2016-08-26 | 2018-03-01 | Netapp, Inc. | Multiple node repair using high rate minimum storage regeneration erasure code |
| CN111818124A (en) * | 2020-05-29 | 2020-10-23 | 平安科技（深圳）有限公司 | Data storage method, data storage device, electronic equipment and medium |
Non-Patent Citations (2)
| Title |
|---|
| JI-GUANG WANG等: "一种采用纠删码缓存的高能效且可靠的存储系统（英文）", 《FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING》 * |
| XIN XIE等: "AZ-Recovery: An Efficient Crossing-AZ Recovery Scheme for Erasure Coded Cloud Storage Systems", 《 2020 INTERNATIONAL SYMPOSIUM ON RELIABLE DISTRIBUTED SYSTEMS (SRDS)》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114595092A (en) * | 2022-04-28 | 2022-06-07 | 阿里云计算有限公司 | Distributed storage system, data reconstruction method, device and storage medium |
| CN114595092B (en) * | 2022-04-28 | 2022-09-20 | 阿里云计算有限公司 | Distributed storage system, data reconstruction method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112783688B (en) | 2022-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11614867B2 (en) | Distributed storage system-based data processing method and storage device | |
| CN106776130B (en) | Log recovery method, storage device and storage node | |
| US11307776B2 (en) | Method for accessing distributed storage system, related apparatus, and related system | |
| RU2501072C2 (en) | Distributed storage of recoverable data | |
| CN106708653B (en) | Mixed tax big data security protection method based on erasure code and multiple copies | |
| EP2908254A1 (en) | Data redundancy implementation method and device | |
| CN104935654A (en) | Caching method, write point client and read client in server cluster system | |
| CN109582213B (en) | Data reconstruction method and device and data storage system | |
| JP2016513306A (en) | Data storage method, data storage device, and storage device | |
| CN109726036B (en) | Data reconstruction method and device in storage system | |
| CN111309245B (en) | Hierarchical storage writing method and device, reading method and device and system | |
| US7849355B2 (en) | Distributed object sharing system and method thereof | |
| CN104486438A (en) | Disaster-tolerant method and disaster-tolerant device of distributed storage system | |
| CN115167782B (en) | Temporary storage copy management method, system, equipment and storage medium | |
| CN111046024A (en) | Data processing method, device, equipment and medium for sharing storage database | |
| CN112783445A (en) | Data storage method, device, system, electronic equipment and readable storage medium | |
| CN112783688B (en) | Erasure code data recovery method and device based on available partition level | |
| CN108616598B (en) | Data synchronization method and device and distributed storage system | |
| CN105550230A (en) | Method and device for detecting failure of node of distributed storage system | |
| CN107045426B (en) | Multi-copy reading method and system | |
| US20180217906A1 (en) | Method For Optimizing Reconstruction Of Data For A Hybrid Object Storage Device | |
| JP6376626B2 (en) | Data storage method, data storage device, and storage device | |
| CN103714059A (en) | Data updating method and device | |
| CN115756955A (en) | Data backup and data recovery method and device and computer equipment | |
| CN108174136B (en) | Cloud disk video coding storage method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |